Stability of a dominant sponge-symbiont in spite of antibiotic-induced microbiome disturbance.

Abstract

Marine sponges are known for their complex and stable microbiomes. However, the lack of a gnotobiotic sponge-model and experimental methods to manipulate both the host and the microbial symbionts currently limit our mechanistic understanding of sponge-microbial symbioses. We have used the North Atlantic sponge species Halichondria panicea to evaluate the use of antibiotics to generate gnotobiotic sponges. We further asked whether the microbiome can be reestablished via recolonization with the natural microbiome. Experiments were performed in marine gnotobiotic facilities equipped with a custom-made, sterile, flow-through aquarium system. Bacterial abundance dynamics were monitored qualitatively and quantitatively by 16 S rRNA gene amplicon sequencing and qPCR, respectively. Antibiotics induced dysbiosis by favouring an increase of opportunistic, antibiotic-resistant bacteria, resulting in more complex, but less specific bacteria-bacteria interactions than in untreated sponges. The abundance of the dominant symbiont, Candidatus Halichondribacter symbioticus, remained overall unchanged, reflecting its obligately symbiotic nature. Recolonization with the natural microbiome could not reverse antibiotic-induced dysbiosis. However, single bacterial taxa that were transferred, successfully recolonized the sponge and affected bacteria-bacteria interactions. By experimentally manipulating microbiome composition, we could show the stability of a sponge-symbiont clade despite microbiome dysbiosis. This study contributes to understanding both host-bacteria and bacteria-bacteria interactions in the sponge holobiont.